Industrial machinery fleets in developing economies face significant operational risks, yet systematic, data-driven methodologies for quantifying and mitigating these risks are scarce. Existing approaches often rely on cross-sectional data, failing to capture temporal dynamics and unobserved heterogeneity within fleets. This paper aims to develop and evaluate a methodological framework for the empirical analysis of machinery fleet risk. The primary objective is to apply panel-data econometric techniques to measure the effectiveness of targeted maintenance interventions on risk reduction. A longitudinal dataset was constructed from maintenance logs, incident reports, and operational records for a fleet of heavy industrial equipment. A two-way fixed effects model was employed to control for time-invariant machine-specific factors and common temporal shocks. The core specification is Risk₈ₓ = ᵢ + ₜ + Intervention₈ₓ + X₈ₓ + ₈ₓ, where robust standard errors were clustered at the machine level. The implementation of structured preventative maintenance protocols was associated with a statistically significant 18. 2% reduction in major incident risk (95% CI: 12. 5% to 23. 9%). Unobserved machine heterogeneity accounted for a substantial portion of the variance in baseline risk profiles. Panel-data estimation provides a robust methodological advance for isolating the causal effect of risk-reduction strategies in industrial settings, moving beyond descriptive correlation. Fleet managers should adopt panel-data frameworks for continuous safety performance evaluation. Investment in digitised, time-consistent record-keeping is essential to enable such analyses. reliability engineering, maintenance optimisation, fixed effects model, industrial safety, asset management This paper presents a novel application of econometric panel-data methods to engineering asset management, demonstrating a rigorous approach to quantifying the efficacy of safety interventions in an industrial context.
Kavishe et al. (Thu,) studied this question.